Utilization also can be measured in terms of the amount, or intensity, of services received by those who visit a dentist. The intensity of service often is measured in terms of the number of visits per person within a specified time interval (such as visits per year) or expenditures per person. As with the percentage of the population that had had at least one dental visit, a previous study also has shown that the number of visits and the mean expenditure per patient also vary by such factors as income, race, sex and educational level.²

Although important and informative, these measures of dental use and intensity can be refined further. For instance, utilization can be explained further with information about the type of service received. In a similar fashion, expenditures also can be clarified further. For instance, expenditures can be analyzed in terms of the sum of charges, sum of payments or sources of payments. While sum of payments may help to provide an accurate measure of reimbursement received by providers (that is, gross receipts), the sum of charges may provide a more accurate measure of the total resources used to provide all services rendered. In some instances, such as insurance discounts, charges represent the nondiscounted fee. In other cases, such as free care or bad debt, charges also capture the total value of services provided when the provider is not fully reimbursed.

This article focuses on charges as a measure of dental expenditure. We examine changes in expenditures from 1987 to 1996 among children 2 to 17 years old. To adjust for inflation, dental expenditures (charges) are measured in terms of real, or constant, dollars. We also examine changes in the amount of these services from 1987 to 1996 within major demographic and socioeconomic categories, as well as changes in sources of funding.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The 1987 National Medical Expenditure Survey, or NMES, was designed to provide detailed national estimates of health expenditures, utilization, sources of payment and insurance coverage for the civilian noninstitutionalized population of the United States during the period from Jan. 1 through Dec. 31, 1987. It was sponsored by the Agency for Healthcare Research and Quality, or AHRQ (formerly the Agency for Health Care Policy and Research and, before that, the National Center for Health Services Research). NMES was a survey of approximately 34,459 people in 14,000 households; it oversampled certain population groups, including elderly people, people with limitations in activities of daily living, African-Americans, Hispanics and the poor. Data were gathered in five rounds of interviews over an 18-month period during 1987 and 1988. The combined response rate was 79.7 percent. The 1987 NMES data were released for public use in October 1992.^3,4

The 1996 Medical Expenditure Panel Survey, or MEPS, also was sponsored by AHRQ and was conducted to provide nationally representative estimates of health care use, expenditures, sources of payment and insurance coverage for the U.S. civilian noninstitutionalized population. However, it differs from the 1987 NMES in that data on household respondents in each panel are collected for two consecutive years and the survey is fielded continuously (that is, a new panel is selected every year). The sample for the 1996 MEPS consisted of 21,571 people in 10,500 households participating in the NCHS’s National Health Interview Survey. To collect health expenditure and utilization data for 1996, field researchers interviewed each MEPS household in person three times over an 18-month period. The combined response rate was 70 percent.⁵

All expenditure estimates presented in this article are per capita and are based on people who had had at least one dental visit during the year of the survey. Dental expenditure and source-of-payment information were collected as part of the household survey. In settings that do not normally specify a total charge, AHRQ staff imputed a dollar value to construct a total expense variable. They also designed a series of edits and imputation procedures for missing or inconsistent expense values and addressed three major issues:

– potential copayment amounts;

– potential flat fees;

– missing charges in cases wherein care was delivered free by the provider or under a prepaid arrangement such as a health maintenance organization, was covered by a flat fee or by Medicaid, or was not otherwise reported by the respondent.

AHRQ staff imputed Medicaid expenses using a database consisting of average 1987 or 1996 Medicaid reimbursements for dental care by state, patient age and type of service.

All expenditure estimates contained in the tables and figures in this article are real (base = 1998), to allow for comparisons over time using constant dollars. We used the dental component of the consumer price index to adjust for the effects of inflation. Only one nominal estimate (in other words, current dollars) is presented, at the beginning of the results section, to demonstrate the importance of removing the effects of inflation when making comparisons over time.

In this article, we focus on three main sources of funding—out-of-pocket, private insurance and public sources—and a fourth category called "not reimbursed." Public sources included Medicare, Medicaid and other federal, state and local government sources.

In the 1987 NMES, expenditure estimates were based on charges.^3,4 A category defined as "free from provider, including professional courtesy and bad debt" was included in expenditures; here, we have labeled this category as "not reimbursed." In the 1996 MEPS, unlike the earlier survey, expenditures were defined as the sum of direct payments for care provided during the year, but there also is a variable that captures total charges.⁴ The difference between charges and expenditures in the 1996 MEPS is defined by AHRQ as uncollected liability, bad debt, charitable care and discounting. This difference between charges and expenditures in the 1996 MEPS is considered as "not reimbursed" in this article and is used for purposes of comparisons with the 1987 NMES. In both surveys, this category represents dental services rendered for which the dentist had received no payment. The not-reimbursed category makes up a significant percentage of total dollar value of dental services in both surveys; therefore, we included it to provide a complete picture of the total output of the dental delivery system.

All expenditure estimates presented in this article are per capita and are based on people who had had at least one dental visit during the year of the survey.

To ensure sufficient numbers to produce reliable national estimates, we combined socioeconomic variable categories when necessary. We used the SUDAAN statistical package (Version 7.11, Research Triangle Institute, Research Triangle Park, N.C.) to calculate standard errors and perform statistical tests because it can adjust for the correlation introduced by the complex sample design used for the 1987 NMES and 1996 MEPS surveys.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Per capita dental expenditures among those who had had a dental visit. Average nominal per capita dental expenditures for children 2 to 17 years of age who had had a dental visit increased from $315.21 in 1987 to $456.98 in 1996. However, when the effect of inflation is removed from these numbers, the NMES and MEPS data show a decrease of 13.7 percent in average real expenditures, from $578.05 to $498.57 (Table 1).

View this table: [in this window] [in a new window]   TABLE 1 1987 NMES* AND 1996 MEPS REAL (BASE = 1998) EXPENDITURE DATA FOR CHILDREN 2 TO 17 YEARS OF AGE WITH A DENTAL VISIT.

Expenditures among white children who visited a dentist were higher, on average, than those among black children in both 1987 and 1996. Among white children, expenditures were 18.3 percent less, on average, in 1996 than in 1987. The absolute difference between white and black children fell from $315.14 in 1987 to $157.66 in 1996. The same pattern was seen for non-Hispanic vs. Hispanic children. The absolute difference between these two groups fell from $245.19 in 1987 to $71.67 in 1996.

Real expenditures increased by 83.9 percent among children in households with incomes below the federal poverty level, or FPL, from 1987 to 1996, from $219.49 to $403.73. However, expenditures decreased by 17.7 percent for children living above the FPL, from $624.20 to $513.77.

Real dental expenditures tended to increase with the child’s level of household income (that is, percentage of poverty level) in both surveys. However, the absolute difference between the highest and lowest income levels fell from $556 in 1987 to $230 in 1996 (Figure 1). In 1996, children below the FPL who visited a dentist reported higher expenditures for dental care, on average, than children in the 100 to 200 percent of FPL income category—$403.73 vs. $275.77.

View larger version (39K): [in this window] [in a new window]   Figure 1. 1987 National Medical Expenditure Survey, or NMES, and 1996 Medical Expenditure Panel Survey, or MEPS, real (base = 1998) dental expenditures by income level among children 2 to 17 years old with a dental visit. FPL: Federal poverty level.

View this table: [in this window] [in a new window]   TABLE 2 1987 NMES* AND 1996 MEPS REAL (BASE = 1998) EXPENDITURE DATA FOR CHILDREN 2 TO 17 YEARS OF AGE WITH A DENTAL VISIT, BY AGE AND POVERTY LEVEL.

Much of the difference in expenditures on dental services related to income among children in this age group is related to orthodontic services. When these services are excluded, the difference in dental expenditures between the highest and lowest income groups is reduced to about $86 in 1996 ($362 – $276) (Figure 2).

View larger version (61K): [in this window] [in a new window]   Figure 2. 1996 Medical Expenditure Panel Survey, or MEPS, dental expenditures by income level, with orthodontic care vs. without orthodontic care, among children 5 to 17 years old with a dental visit. FPL: Federal poverty level.

View this table: [in this window] [in a new window]   TABLE 3 1987 NMES* AND 1996 MEPS REAL (BASE = 1998) PER CAPITA EXPENDITURE DATA FOR CHILDREN 2 TO 17 YEARS OF AGE WITH A DENTAL VISIT, BY POVERTY AND SOURCES OF FUNDING.

View larger version (41K): [in this window] [in a new window]   Figure 3. 1996 Medical Expenditure Panel Survey, or MEPS, sources of funding for dental services by income level among children 2 to 17 years old with a dental visit. Pocket: Out-of-pocket funding. Private: Private insurance. Public: Public funding.

Decreases in real per capita expenditures among the top two income categories were related mostly to a decrease in out-of-pocket payments.

Overall, out-of-pocket payments declined in real dollars from $303.12 in 1987 to $189.28 in 1996. As a percentage of total funding, out-of-pocket payments fell from 52.4 percent in 1987 to 38.0 percent in 1996. The contributions from private insurance and public funding declined in real dollars, but held steady in terms of percentage of total funding. Private insurance accounted for 37.4 percent of total funding in 1996 and public funding accounted for 4.4 percent. The not-reimbursed category grew from $27.64 to $93.47, or from 4.8 percent of the total in 1987 to 18.7 percent of the total in 1996.

For children below the FPL, the real increase in funding from $219.49 in 1987 to $403.72 in 1996 was due in large part to an increase in the "not reimbursed" category, which rose from 6.2 percent of the total in 1987 to 40.7 percent of the total in 1996 (Figure 4, page 480). Out-of-pocket payments for these children fell from 50.4 percent of the total to 23.9 percent in 1996. Public funding doubled in terms of real dollars, from $57.45 to $114.12. As a percentage of the total, the increase was from 26.2 percent to 28.3 percent.

View larger version (44K): [in this window] [in a new window]   Figure 4. 1987 National Medical Expenditure Survey, or NMES, and 1996 Medical Expenditure Panel Survey, or MEPS, sources of funding for dental services by income level among children 2 to 17 years old with a dental visit. FPL: Federal poverty level. Pocket: Out-of-pocket funding. Private: Private insurance. Public: Public funding.

Decreases in real per capita expenditures among the top two income categories were related mostly to a decrease in out-of-pocket payments. Despite an overall decrease in expenditures, there was an increase in the not-reimbursed category, both in terms of amount and percentage of the total, from 1987 to 1996.

Finally, we have discussed results presented in this article based on sources of funding in terms of per capita average amounts and as a percentage of total funding. Table 4 (page 481) presents an estimate of the total dollar value of not-reimbursed care provided to children 2 to 17 years of age by income level based on the 1996 MEPS. Also included in Table 4 is an estimate of the total dollar value of dental care provided to children 2 to 17 years of age in 1996 by income level.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
For children who visited a dentist, these findings indicate a decline in real per capita expenditures. Overall, among children aged 2 to 17 years, real per capita expenditures (base = 1998) declined by about 14 percent, from $578 to $499 over the period from 1987 to 1996. Dental expenditures among children 5 to 17 years old with a visit were less on average in 1996 than in 1987. Children aged 2 to 5 years who had had a dental visit did report an increase, but this increase was from a relatively small base in 1987.

Racial and ethnic disparities in expenditures among children who visited the dentist were dramatically reduced from 1987 to 1996. Comparable reductions in disparities due to income also were found.

Large expenditure increases per patient occurred among the poorest children, while expenditure decreases occurred among children from families with higher incomes. The reasons for these results are not completely understood. A possible explanation may lie in the decrease in untreated caries among children. Children in higher-income households have low levels of cumulative caries (both treated and untreated) and almost no untreated caries, and they have had those low levels for some time.^6–8 Thus, middle-class children did not need much restorative dentistry. Expenditures for these children largely were for routine diagnostic and preventive services and did not increase between 1987 and 1996. Lower-income children also showed a marked decrease in both cumulative and untreated caries. Nevertheless, these children had a greater need for restorative care, even after the decrease. These findings suggest that lower-income children received more of the care they needed in 1996 than in 1987.

Among children 5 to 17 years of age, most of the gap in expenditures between the highest and lowest income groups in 1996 was due to orthodontic services. When expenditures for orthodontic services were excluded from the analysis, the gap between expenditures between the lowest income and highest income group was reduced to $86.36 in 1996.

Children below the FPL still are less likely to visit a dentist than their upper-income counterparts.⁹ Nevertheless, once a visit has occurred, it appears that children at different income levels receive similar amounts of care as measured by dental expenditures.

Among children living at 100 to 200 percent of the FPL, real per capita expenditures fell from 1987 to 1996. This is a disturbing trend. The roles played by different funding sources may be part of the explanation. The amount of care paid for by two important sources of funding, out-of-pocket payments and public funding, decreased both in amount and as a percentage of the total during this period. Funding from private insurance declined in amount, but its percentage contribution remained level. Dentists’ care that was "not reimbursed" increased in both amount and as a percentage of the total, but could not make up for the declines in other sources of funding.

It should be noted that this trend occurred before the State Children’s Health Insurance Program, or SCHIP, was enacted in 1997. We can hope that this program will help reverse this trend in expenditures and increase the amount of dental services these children receive. For that to happen, SCHIP must be funded at adequate levels, and dentists must be adequately reimbursed for their services.

A possible unintended consequence of SCHIP that should be followed closely is the possibility that increased public funding for children eligible for SCHIP could result in a decline in private insurance coverage as children become covered by SCHIP. This would represent a shift from private funding to public funding. To the extent that this occurs, the amount of additional funding for these children would be reduced by the amount that public funding is substituted for private funding.

Nonreimbursed dental care provided to children by dentists is substantial, both in amount and as a percentage of all dental care. In 1996, nonreimbursed care amounted to $2.6 billion of $13.8 billion total, or about 19 percent of all children’s dental care. In 1987, this percentage was 4.8 percent.

The reasons for this increase cannot be demonstrated with these data. Within the not-reimbursed category, it is not possible to determine the individual contributions of charity care, bad debt, professional courtesy, uncollected liability and discounting. These may vary by the income level of the child receiving the care. Charity care may be more common among lower-income children and discounting more common among higher-income children with dental insurance. In addition, it is not possible to determine if some of the difference may be due to measurement error or differences in design between the two surveys.¹⁰ On the other hand, the increase in this category from 1987 to 1996 is notable, and this category represented a significant portion of care received, charges incurred and resources used by these patients. Consequently, it is important to include the not-reimbursed category as part of these analyses.

An examination of this category shows that the amount not reimbursed may vary by the income level of the child receiving the care. About two-thirds of the total dollar value of "not reimbursed" care went to children in the top two income categories in 1996. However, children from lower-income families also received a substantial amount of this care. Dentists provided, without reimbursement, a significant percentage (40.7 percent of per capita funding) of all dental care that below-poverty-level children received in 1996. This represents an important contribution by practicing dentists toward maintaining the oral health of these children.

	CONCLUSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Overall, real per-patient dental expenditures among children 2 to 17 years old fell from $578 in 1987 to $499 in 1996. Racial and ethnic disparities in dental expenditures were dramatically reduced from 1987 to 1996. In terms of sources of funding, out-of-pocket payments fell from 52.2 percent of the total to 38 percent, contributions from private insurance and public funding held steady at 37 percent and 4.5 percent of the total, respectively, while the not-reimbursed category grew from 4.8 percent to 18.7 percent of the total.

Real per-patient dental expenditures among children living in households below the FPL increased from $219 in 1987 to $404 in 1996. While this finding provides the basis for optimism regarding access to dental care for lower-income children, more can and should be done to increase the percentage of these children who visit a dentist. All segments of society must cooperate to achieve this result. The political will must be found to increase public funding for this important segment of our population.

View larger version (91K): [in this window] [in a new window]   Mr. Wall is the manager, Statistical Research, Health Policy Resources Center, American Dental Association, Chicago.

View larger version (88K): [in this window] [in a new window]   Dr. Brown is the associate executive director, Health Policy Resources Center, American Dental Association, 211 E. Chicago Ave., Chicago, Ill. 60611, e-mail "brownja@ada. org". Address reprint requests to Dr. Brown.

View larger version (91K): [in this window] [in a new window]   Dr. Manski is a senior scholar, Agency for Healthcare Research and Quality, Rockville, Md. He also is a professor, Department of Oral Health Care Delivery, Dental School, University of Maryland, Baltimore.